Re: SendRowDescriptionMessage() is slow for queries with a lot of columns

From: tushar <tushar(dot)ahuja(at)enterprisedb(dot)com>
To: Andres Freund <andres(at)anarazel(dot)de>, pgsql-hackers(at)postgresql(dot)org
Subject: Re: SendRowDescriptionMessage() is slow for queries with a lot of columns
Date: 2017-09-29 09:02:30
Message-ID: 1361a409-c997-ac38-8374-9fba944938c7@enterprisedb.com
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-hackers

On 09/27/2017 10:50 PM, Andres Freund wrote:
> This'll allow the later patches to allow the compiler to perform the
> relevant optimizations. It also allows to optimize e.g. pq_sendint64()
> to avoid having to do multiple byteswaps.

After applying all the required patches, able to see some performance gain

Virtual Machine configuration - Centos 6.5 x64 / 16 GB RAM / 8 VCPU core
processor

./pgbench -M prepared -j 10 -c 10 -f /tmp/pgbench-many-cols.sql postgres
-T TIME

After taking Median of 3 run  -

Case 1 – TIME=300

PG HEAD =>41285.089261 (excluding connections establishing)
PG HEAD+patch =>tps= 42446.626947(2.81+% vs. head)

Case 2- TIME=500

PG HEAD =>tps = 41252.897670 (excluding connections establishing)
PG HEAD+patch =>tps= 42257.439550(2.43+% vs. head)

Case 3- TIME=1000

PG HEAD =>tps = 1061.031463 (excluding connections establishing)
PG HEAD+patch => tps= 8011.784839(3.30+% vs. head)

Case 4-TIME=1500

PG HEAD =>tps = 40365.099628 (excluding connections establishing)
PG HEAD+patch =>tps= 42385.372848(5.00+% vs. head)

--
regards,tushar
EnterpriseDB https://www.enterprisedb.com/
The Enterprise PostgreSQL Company

In response to

Responses

Browse pgsql-hackers by date

  From Date Subject
Next Message Maksim Milyutin 2017-09-29 09:18:28 Re: Partitions: \d vs \d+
Previous Message Craig Ringer 2017-09-29 08:27:58 Re: pg_prepared_xact_status